A major industrial challenge comprises the development of efficient methods for reducing the concentration of air pollutants, such as sulfur oxides, in waste gases, such as waste gases resulting from the processing and combustion of sulfur-containing hydrocarbon fuels. The discharge of these waste gas streams into the atmosphere is environmentally undesirable at the sulfur oxide concentrations which are frequently encountered in conventional operations. Such waste gas streams typically result, for example, from the combustion of sulfur-containing fossil fuels for the generation of heat and power, the regeneration of catalysts employed in the refining of hydrocarbon feedstocks which contain organic sulfur compounds, and the operation of Claus-type sulfur recovery units.
Two fundamental approaches have been suggested for the removal of sulfur oxides (SO.sub.x) from a waste gas. One approach involves scrubbing the waste gas with an inexpensive alkaline material, such as lime or limestone, which reacts chemically with SO.sub.x yielding a non-volatile product for disposal. Unfortunately, this approach requires a large and continual supply of the alkaline scrubbing material, and the resulting reaction products can create a solid waste disposal problem of substantial magnitude. The second principal approach to the control of SO.sub.x emission involves the use of SO.sub.x absorbents which can be regenerated either thermally or chemically.
Numerous materials have been proposed for use in removing SO.sub.x from gases. For example, Bertolacini et al., U.S. Pat. No. 3,835,031, disclose the use of a crystalline aluminosilicate cracking catalyst impregnated with a Group IIA metal compound or mixture thereof as an oxide or oxides, inclusive of magnesium oxide or magnesia (MgO), for reduced SO.sub.x emission in the regenerator stack gases.
De'Souza et al., U.S. Pat. No. 4,233,276, disclose a method for removing oxidizable sulfur compounds from a waste gas utilizing a metal oxide absorbent, inclusive of sodium, potassium, lithium, magnesium, calcium, strontium, barium, scandium, titanium, chromium, iron, molybdenum, manganese, cobalt, nickel, antimony, copper, zinc, cadmium, lead and rare earth metals, and further inclusive of oxidation promoters such as rutheniuam, osmium, rhodium, silver, iridium, palladium, platinum, vanadium and molybdenum.
Bertolacini et al., U.S. Pat. No. 4,369,130, disclose a fluidized catalytic cracking (FCC) catalyst in combination with an absorbent rare earth metal compound and an inorganic oxide, such as an oxide of aluminum, magnesium, zinc, titanium and calcium. The disclosed absorbent can be circulated through an FCC system together with the hydrocarbon cracking catalyst to reduce Sox emissions from the regenerator zone.
Bertolacini et al., U.S. Pat. No. 4,381,991, disclose a process for removing SO.sub.x from a gas using an absorbent comprising MgO in combination with at least one rare earth metal. The disclosed absorbent can be circulated through an FCC system together with the hydrocarbon cracking catalyst to reduce SO.sub.x emissions from the regenerator zone.
Lewis et al., U.S. Pat. No. 4,626,419, disclose a process for removing SO.sub.x from a mixture of gases utilizing a composite containing a porous refractory support bearing a first component comprising bismuth, chromium or a rare earth metal, such as cerium, and a second component comprising an alkali metal, such as potassium. Burk, Jr. et al., U.S. Pat. No. 4,735,705, disclose an FCC process employing an FCC catalyst in combination with SO.sub.x absorbent particles comprising at least one spinel containing an additive inclusive of an alkali metal, calcium, barium, strontium, beryllium and mixtures thereof. Dai et al., U.S. Pat. No. 5,021,228, disclose the removal of SO.sub.x from gases by contact with a composition comprising alumina containing potassium and thorium. Magnabosco et al., U.S. Pat. No. 5,108,979, disclose the use of spinels to reduce levels of SO.sub.x in the context of an FCC system.
Kim, U.S. Pat. No. 5,288,675, discloses an SO.sub.x gettering composition for use in an FCC system. The disclosed gettering composition comprises an attrition-resistant, coprecipitated magnesia-lanthana-alumina component combined with a catalytic oxidation and/or reducing promoter metal such as ceria, vanadia and/or titania.
Buchanan et al., U.S. Pat. No. 5,547,648, disclose a method of removing SO.sub.x from a combustion flue gas stream emitted from an FCC reactor utilizing an absorbent comprising any of numerous components, such as Group IA metals, Group IIA metals, and Group VIII metals. Magnesium aluminate spinels impregnated with vanadium and cerium are disclosed as particularly useful.
Moore et al., European Patent Application No. EP 0 247 836 A1 disclose an FCC catalyst inventory comprising an FCC catalyst and an SO.sub.x absorbent comprising one or more rare earth metal oxides, particularly lanthanium or cerium, supported on attrition-resistant particles of alumina or a magnesia-alumina spinel. Moore et al. recognize that the limited commercial success of various SO.sub.x additives is due to the exigencies of an FCC system.
The cyclic, FCC of heavy petroleum fractions is one of the major refining operations involved in the conversion of crude petroleum oils to valuable products, such as the fuels utilized in internal combustion engines. A typical FCC unit comprises three sections: a cracking section or reactor; a regenerator and a separation section or stripping zone. A typical FCC process involves continuous catalytically cracking of a petroleum feedstock in a reactor zone through contact with a particulate FCC catalyst at temperatures between about 400.degree. C. and about 700.degree. C. Particulate FCC catalysts substantially deactivated by non-volatile, sulfur-containing coke deposits are separated from the reactor zone effluent and stripped of volatile deposits in a stripping zone. The stripped catalyst particles are separated from the stripping zone effluent, regenerated in a regenerator zone by combustion of the coke with an oxygen-containing gas at temperatures between about 565.degree. C. and about 790.degree. C., and the regenerated catalyst particles returned to the reactor zone. The combustion of sulfur-containing coke results in the release of substantial amounts of SO.sub.x to the atmosphere.
While numerous materials and composites are known to have absorbent and catalytic properties in connection with SO.sub.x reduction, the formulation of an efficient SO.sub.x reducing additive, e.g., catalyst and/or absorbent, in the context of an FCC system and it exigencies is fraught with problems and unpredictability.
Generally, about 45% to about 55% of the sulfur in the hydrocarbon feedstock is converted to hydrogen sulfide (H.sub.2 S) in the FCC reactor, about 35% to about 45% remains in a liquid product, and about 5% to about 10% in the coke deposited on the FCC catalyst. These amounts vary depending upon the type of hydrocarbon feedstock, rate of hydrocarbon cycle, steam stripping rate, type of FCC catalyst, reactor temperature, reactor design and other FCC system variables. Accordingly, the formulation of an effective additive for reducing SO.sub.x emissions from an FCC system is recognized in the art as a challenging problem. See the previously mentioned Moore et al., EP 0 247 836 A1; and Bhattacharyya et al., "Catalytic SO.sub.x Abatement: The Role of Magnesium Aluminate Spinel in the Removal of SO.sub.x from Fluid Catalytic Cracking (FCC) Flue Gas," Ind. Eng. Chem. Res. 1988, 27, pp. 1356-1360.
The difficulties attendant upon formulating and designing an effective SO.sub.x reducing additive in the context of an FCC system stems from various requirements and considerations, aside from the generally unpredictable nature of catalytic activity. The particulate material serving as the SO.sub.x reducing additive must be attrition-resistant to survive in an FCC environment without fragmenting. Accordingly, an effective SO.sub.x reducing additive should have a Davison Index less than 10. An effective particulate SOX reducing additive should not contain any metal or other component which actually functions as a poison in the FCC regime. In addition, an effective particulate SOX catalyst/absorbent must perform three functions: (1) oxidize SO.sub.2 to SO.sub.3 ; (2) chemisorb SO.sub.3 ; and (3) release the absorbed SO.sub.3 as H.sub.2 S in the reactor side of an FCC system. During regeneration, sulfur in the coke is oxidized primarily to SO.sub.2. In order for sulfate chemisorbtion to occur, the SO.sub.2 must be oxidized to SO.sub.3 which is then chemisorbed as the sulfate. As the operational temperature of the regenerator is increased, the formation of SO.sub.3 is less favored. Accordingly, the catalyzing function of an SO.sub.x catalyst/absorbent is significant.
In FCC units operating with high sulfur-containing feedstocks, relatively large amounts of sulfur acceptors having a high unit capacity to adsorb SO.sub.x are required to accomplish reductions in SO.sub.x levels. The use of large amounts of an SO.sub.x reducing additive results in appreciable dilution of the active FCC catalyst in the cracking reaction cycle whether the sulfur acceptor is a part of the FCC particle itself or is present as a discrete entity circulated with the FCC catalyst inventory. A basic limitation is that conditions of time and temperature for operating cyclic FCC units, especially heat balanced FCC units, are geared to maximizing the production of desired products. Conditions established to achieve this result are by no means those that are optimum for reversibly reacting SO.sub.x in the regenerator zone and carrying the sulfur back to the reactor for conversion at least in part to H.sub.2 S. Although SO.sub.x reducing additives offer promise, they leave much to be desired because, inter alia, SO.sub.x removal activity decreases rapidly with the residence time available for such SO.sub.x reducing additives to function effectively.
Desorption of absorbed SO.sub.x typically involves hydrolysis of a metal sulfide to form the original metal oxide yielding H.sub.2 S. Therefore, an effective SO.sub.x catalyst/absorbent must be capable of liberating the absorbed sulfur in the form H.sub.2 S under conditions prevailing in the reactor portion of an FCC system. Bhattacharyya et al., in the previously mentioned publication, reported the results of experimental testing to determine the feasibility of actually employing various SO.sub.x catalytic/absorbents in an FCC unit. Among the candidates studied were MgO impregnated with ceria (CeO.sub.2) for oxidizing SO.sub.2 to SO.sub.3, and a magnesium aluminate spinel (Mg.sub.2 Al.sub.2 O.sub.5). Vanadium pentoxide (V.sub.2 O.sub.5) was recognized as an excellent oxidation catalyst for converting SO.sub.2 to SO.sub.3. However, V.sub.2 O.sub.5 was shunned because of its expected undesirable reaction with zeolites, the predominant type of FCC catalyst. Accordingly, CeO.sub.2 was employed for the oxidation of SO.sub.2 to SO.sub.3. Another tested catalyst was CeO.sub.2 on gamma alumina.
The testing reported by Bhattacharyya et al. reveals that CeO.sub.2 in gamma alumina was not a very effective SO.sub.x catalyst/absorbent. CeO.sub.2 impregnated MgO was found to be significantly more effective in absorbing SO.sub.3 than CeO.sub.2 on gamma alumina. However, magnesium sulfate (MgSO.sub.4) is extremely stable and, hence, could not be regenerated as efficiently as the CeO.sub.2 -gamma alumina catalyst/absorbent. Testing revealed that about 27% of absorbed material remained with the CeO.sub.2 --MgO SO.sub.x catalyst/absorbent, even after 20 minutes of hydrogen reduction, possibly as magnesium sulfide (MgS) or unreduced MgSO.sub.4. Bhattacharyya et al. concluded that the rapid deactivation of the CeO.sub.2 --MgO SO.sub.x catalyst-absorbent is a major reason why it was not considered as a potential SO.sub.x catalyst/absorbent for FCC systems. In addition, MgO lacks the requisite attrition-resistant properties for FCC application, in that it is very soft and breaks down into very fine particles in a short period of time. The unsuitability of MgO as an SO.sub.x catalyst/absorbent in an FCC environment is apparently well known in the art and also reported by Magnabosco et al. in U.S. Pat. No. 5,108,979, wherein the presence of free magnesium oxide is disclosed as undesirable. The experimental testing conducted by Bhattacharyya et al. led to the conclusion that thermally stable magnesium aluminate spinels, such as MgAl.sub.2 O.sub.4 or Mg.sub.2 Al.sub.2 O.sub.5, were best suited for FCC application. Indeed, a spinel base catalyst, such as that commercialized as DESOX.TM., is currently recognized as the SO.sub.x catalyst/absorbent of choice for FCC systems.
Copending application Ser. No. 08/874,516, filed on Jun. 13, 1997, discloses an SO.sub.x catalyst/absorbent comprising MgO, CeO.sub.2 and/or Ag, and V.sub.2 O.sub.5. The disclosed SO.sub.x catalyst/absorbent rapidly and effectively releases absorbed SO.sub.x and enjoys utility in an FCC system, as part of the catalyst inventory.
There is a continuing need for an SO.sub.x reducing additive, suitable for use in an FCC system, which is capable of absorbing SO.sub.x in the regenerator zone, given regenerator zone exigencies, including residence time and temperature, and capable of rapidly releasing the absorbed SO.sub.x in the reactor zone of an FCC system, given reactor zone exigencies, including residence time and temperature. There further exists a need to produce such an SO.sub.x reducing additive in a cost-effective and efficient manner.